I'm wondering what the practicality would be of building ocean bound nuclear power plants to react and create Methanol. Just looking at it. Sea water near the surface has a higher concentration of CO2 in it that the air does, so it might be easier to separate it out of sea water rather than air. CO2 + Hydrogen gets you Methanol. Certainly there are some rather steep losses. I don't remember the numbers now, but I'm thinking less than 1/4 of the energy is retained. (I crunched them in another thread a good while back.)
However, on the upside, the risk to human populations in the event of a melt down is very small, because you can just sink the boat. The ocean can absorb a failure or two if these events are not very frequent.
Another upside is ease of cooling. I don't know if the turbine could work with salt water, but I doubt it. So there'd need to be some use of fresh water. Either that or desalinization.
Are any of the "wonder tech's" like Thorium good enough that they could be used to generate power at a price low enough to overcome the energy losses? I mean, if you're losing so much that you only retain 1/4 of your energy, but at the same time the energy itself only costs you 1/5 the normal amount, then you're still making a profit in spite of the losses.